latest version

app.py

@@ -10,7 +10,7 @@ from langchain.text_splitter import TokenTextSplitter, CharacterTextSplitter
 from langchain import PromptTemplate, LLMChain
 from langchain.chains.question_answering import load_qa_chain
 from langchain.vectorstores import FAISS
-from langchain.document_loaders import TextLoader
+from langchain.document_loaders import TextLoader, UnstructuredEPubLoader
 from langchain.callbacks.manager import CallbackManager
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
 import gradio as gr
@@ -42,6 +42,20 @@ llm = HuggingFaceHub(
 embeddings = HuggingFaceEmbeddings()
 callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])
 
+def upload_file(File):
+  if File.name.endswith(".txt"):
+    read_textfile(File)
+  elif File.name.endswith(".epub"):
+    read_epub(File)
+
+def read_epub(File):
+  loader = UnstructuredEPubLoader(File.name)
+  documents = loader.load()
+  text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=10)
+  docs = text_splitter.split_documents(documents)
+  db = FAISS.from_documents(docs, embeddings)
+  db.save_local(folder_path=".", index_name="faiss_index")
+  
 def read_textfile(File):
   loader = TextLoader(File.name)
   documents = loader.load()
@@ -68,7 +82,7 @@ def ask_question(question, chat_history):
 with gr.Blocks() as demo:
   gr.Markdown("## 📚AskYourBooks")
   gr.Markdown("""
-    Ask questions to your books in txt format. After a short indexing time, 
+    Ask questions to your books in txt or epub format. After a short indexing time, 
     ask questions about the content of the book.
     """
   )
@@ -79,7 +93,7 @@ with gr.Blocks() as demo:
   clear_button = gr.Button("Clear conversation")
   chatbot = gr.Chatbot(label="Conversation")
   
-  textfile.change(fn=read_textfile, inputs=[textfile], outputs=[])
+  textfile.change(fn=upload_file, inputs=[textfile], outputs=[])
   ask_button.click(
       fn=ask_question, 
       inputs=[ask_textbox, chatbot], 

requirements.txt

@@ -3,4 +3,6 @@ langchain
 transformers
 sentence_transformers
 tiktoken
-faiss-cpu
+faiss-cpu
+unstructured
+pypandoc